Closed GoogleCodeExporter closed 9 years ago
rem-long/2addr is improperly decompiled as multiplication, instead of modulo. For example: calculateChecksum-BB@0x0 : 0 (00000000) const-wide/32 v8, 1000000 # [1000000.0] 1 (00000006) const-wide/16 v6, 1000 # [1000.0] 2 (0000000a) div-long v2, v10, v8 3 (0000000e) rem-long v4, v10, v8 4 (00000012) div-long/2addr v4, v6 5 (00000014) add-long/2addr v2, v4 6 (00000016) rem-long v4, v10, v6 7 (0000001a) add-long/2addr v2, v4 8 (0000001c) div-long v4, v12, v8 9 (00000020) add-long/2addr v2, v4 10 (00000022) rem-long v4, v12, v8 11 (00000026) div-long/2addr v4, v6 12 (00000028) add-long/2addr v2, v4 13 (0000002a) rem-long v4, v12, v6 14 (0000002e) add-long/2addr v2, v4 15 (00000030) div-long v4, v14, v8 16 (00000034) add-long/2addr v2, v4 17 (00000036) rem-long v4, v14, v8 18 (0000003a) div-long/2addr v4, v6 19 (0000003c) add-long/2addr v2, v4 20 (0000003e) rem-long v4, v14, v6 21 (00000042) add-long v0, v2, v4 22 (00000046) const-wide/16 v2, 4096 # [4096.0] 23 (0000004a) rem-long/2addr v0, v2 24 (0000004c) return-wide v0 Is decompiled as: private static long calculateChecksum(long p10, long p12, long p14) { return ((((((((((p10 / 1000000.0) + ((p10 % 1000000.0) / 1000.0)) + (p10 % 1000.0)) + (p12 / 1000000.0)) + ((p12 % 1000000.0) / 1000.0)) + (p12 % 1000.0)) + (p14 / 1000000.0)) + ((p14 % 1000000.0) / 1000.0)) + (p14 % 1000.0)) * 4096.0); } When it should be "% 4096".
Original issue reported on code.google.com by drspring...@gmail.com on 18 Jan 2013 at 3:39
drspring...@gmail.com
Original comment by anthony....@gmail.com on 18 Jan 2013 at 3:44
anthony....@gmail.com
Original comment by Lir...@gmail.com on 18 Jan 2013 at 8:33
Lir...@gmail.com
Original issue reported on code.google.com by
drspring...@gmail.com
on 18 Jan 2013 at 3:39